Development of a Pressure Ulcer Trigger Tool for the Neonatal Population

J Wound Ostomy Continence Nurs. 2012;XX(X):1-5. Published by Lippincott Williams & Wilkins WOUND CARE Development of a Pressure Ulcer Trigger Tool for the Neonatal Population Bette Schumacher Mary Askew Kathy Otten A large Midwest level IIIb neonatal intensive care unit located in a 500-bed teaching hospital implemented quarterly skin prevalence surveys to monitor prevalence of altered skin integrity including pressure ulcers, diaper dermatitis (incontinenceassociated dermatitis), and skin damage as a result of intravenous therapy, adhesive, or medical devices. Pressure ulcer prevalence varied from 0% to 1% per quarter, and no pressure ulcer risk assessment tool was regularly implemented. Therefore, a working group was formed to identify a risk assessment in place. The Iowa Model for Evidence-Based Practice was used to guide the project. A literature review was completed to identify validated instruments, but available tools were judged lengthy for routine clinical use. Therefore, we developed a short trigger tool comprising 3 questions to identify infants at risk for pressure ulcer development. Introduction In 2007, the Centers for Medicare & Medicaid Services defined a number of conditions, including development of hospital-acquired pressure ulcers (PUs) as never events. 1 In response to these challenges, the Institute for Clinical Systems Improvement (ICSI) and the Institute for Healthcare Improvement have published clinical practice guidelines or protocols designed to assist organizations in implementing best practices related to the prevention of PUs. 2, 3 Our level IIIb NICU had no formalized PU risk assessment for bedside nurses. The Iowa Model for Evidence- Based Practice 4 was used to write pertinent evidence-based questions, using the Patient/Intervention/Comparative Intervention/Outcome format. We then reviewed and evaluated pertinent literature, including ICSI and Institute for Healthcare Improvement guidelines, for possible practice changes related to the assessment of PUs in the NICU. The Patient/Intervention/Comparative Intervention/Outcome question and process we identified using the Iowa Model for Evidence-Based Practice are summarized in Table 1. Pressure ulcer development is attributed multiple factors, and a number of tools have been developed and vali- dated in order to evaluate risk in adults. 5-7 Historically classified as occurring almost exclusively in adults, a recent integrative review article described PU occurrence in infants and children. 8 In addition, several studies have examined PU prevalence and incidence rates in the pediatric and neonatal population. 9, 10 For example, Noonan and colleagues 9 conducted a 1-day prevalence survey at tertiary care university-affiliated children's hospital. High risk was noted in 6% of the patients surveyed and a PU prevalence rate of 1.6% was reported. Huffines and Lodgson 10 reported a 19% PU incidence rate in a NICUbased on measurement over a 3-month period. Risk Assessment in the Pediatric and Neonatal Population Given the occurrence of PU in infants and children, a risk assessment tool has been developed for the pediatric population. Similar to Braden's group, 7 Curley and associates 11 grouped risk factors into 2 broad constructs: intensity and duration of pressure, and tolerance of the skin and supporting structure to design the Braden Q. The scale was adapted from the original Braden Scale for Pressure Sore Risk to reflect the unique needs of children, the prevalence of gastric/transpyloric tube feedings, and the availability of blood studies and noninvasive technology in the acutecare pediatric setting. In contrast to the Braden Scale, the Braden Q added an additional risk factor, tissue perfusion, and oxygenation, to the broader construct of tolerance of the skin and supporting structures. Thus, the Braden Q is Bette Schumacher, MS, RN, CPN, Sanford USD Medical Center, Sioux Falls, South Dakota. ; [AQ1] Mary Askew, BSN, RN, CWON, University of Minnesota Medical Center, Minneapolis. ; Kathy Otten, BSN, RN, CWOCN, Sanford USD Medical Center, Sioux Falls, South Dakota.. The authors declare no conflict of interest. Correspondence: Bette Schumacher, MS, RN, CPN, Sanford USD Medical Center, Sioux Falls, SD 57110. [AQ2] DOI: 10.1097/WON.0b013e31826a4d99 Copyright 2012 by the Wound, Ostomy and Continence Nurses Society J WOCN November/December 2012 1 WON200423.indd 1

2 Schumacher et al J WOCN November/December 2012 TABLE 1. Use of the Iowa Model for Evidence-Based Practice Projects to Establish Clinical Tool for Evaluation of Pressure Ulcer Risk in a NICU 1. Generate the question from either a problem or new knowledge 2. Determine relevance to organizational priorities 3. Develop a team to gather and appraise evidence 4. Determine if the evidence answers the question 5. If there is sufficient evidence, pilot the change in practice. 6. Evaluate structure, process, and outcome data For infant in the NICU, does the use of a pressure ulcer trigger tool perform equally as well as the Braden Q to identify infant at risk? P Infant in the NICU I Use of a pressure ulcer trigger tool C Usual practices of assessing all with Braden Q O Trigger tool performs equally as well as the Braden Q for risk identification Our hospital is committed to safe and reliable care and ensuring a flawless patient experience. This includes preventing hospital-acquired pressure ulcers a never event. Possible impact/outcomes may include: Potential increase in WOC referrals Correct triggering of infants requiring further pressure ulcer risk assessment and prevention strategies by nursing Preservation of nursing time since the trigger tool is a shorter and easier-to-use tool that indicates for whom full assessment is needed. Team members included a clinical nurse specialist, 2 WOC nurses, NICU Nursing Practice Council members, and electronic medical records experts. According to IHI, all premature infants are at risk for pressure ulcer development. While this statement is visionary, it does not assist the bedside nurse to determine for whom to provide interventions. Referrals to the WOC nurse for assessment were based on clinical judgment, and no assessment tools were in place. Risk assessment tools are available for the neonatal population but were judged to be lengthy and time-consuming for every nurse, every shift. A trigger tool was developed, based on IHI pediatric trigger questions to help the nurse determine for whom to refer for full assessment. Reasonable evidence is present to warrant implementation of this practice. The 3 IHI trigger questions are based on the concepts of the Braden Q and could trigger those at risk and requiring further assessment and intervention. Prior to implementation, 10 patients were randomly selected to test the feasibility of implementing trigger questions. The 3 trigger questions were asked of the nurses caring for each patient, and results were compared to a Braden Q score generated by a WOC nurse. We noted that patients with high risk as defined by the Braden Q were also identified as at-risk based on the 3 trigger questions. The 3 trigger questions were imbedded into the electronic medical record. Pressure ulcer data were collected through the surveillance provided by skin team nurses as part of their usual duties. The HAPU rate for the NICU remains low and has not changed since the implementation of the trigger tool. The number of WOC nurse consultations has remained stable since implementation of the 3 trigger questions into NICU nurse practice. 7. Disseminate results Results and appropriate feedbacks have been shared with the NICU nurses via the NICU practice committee. Abbreviations: HAPU, hospital-acquired pressure ulcer; IHI, Institute for Healthcare Improvement; NICU, neonatal intensive care unit. composed of 7 subscales that are rated from 1 (worst score) to 4 (best score). The total scale scores range from 7 (indicating the very highest risk) to 28 (lowest risk), and scores 16 or fewer indicate an increased risk for PU development. The scale was validated in 2003 in a multisite prospective cohort descriptive study of 322 pediatric intensive care patients and identified the critical cutoff score. 11 Using the cut score of 16, the sensitivity of the Braden Q was 88% and its specificity was 58%. Although this is a valid and reliable tool, we elected not to use it for routine use by bedside staff because we felt it was too long. Only 1 tool was located in the literature that attempted to quantify PU risk in the neonate. 10 The Neonatal Skin Risk Assessment Scale is also based on the Braden Scale. 7 Three of the 6 subscales were found to have interrater reliability: general physical condition (the infant's gestational age), activity (the type of bed the infant was placed in), and nutrition. Its sensitivity was reported as 83% and specificity was 81%, based on the 3 subcales. 10 We also evaluated the Neonatal Skin Risk Assessment Scale for use in our NICU but elected not to use it because the bed types measured in the instrument are not consistent with our NICU environment, and we deemed the tool to be too long for use in daily clinical practice. Other authors have also stated that risk assessment tools tend to be lengthy and time-consuming for routine clinical WON200423.indd 2

J WOCN Volume xx/number x Schumacher et al 3 use. For example, Anthony and colleagues stated, The complexity of [pressure ulcer] risk assessment scales does not appear to be warranted risk assessment scales are useful research tools, but may not be useful in clinical practice. 12 Most recently, Kottner and colleagues 13 noted no [pressure ulcer risk] instrument [for infants and children] is superior to another maybe clinical judgement is more efficient in evaluating risk than the application of risk scale scores. 13 Nevertheless, we wished to identify or develop a method to accurately and quickly identify neonates at greatest risk for PU development in order to determine how to best allocate limited nursing resources such as staff time, referrals to WOC nurses, specialty equipment, and supplies. In order to fill this clinical need, we elected to develop an instrument that used trigger questions to aid nurses in recognizing PU risk factors in our critically ill neonatal population. Infants who were identified as having risk factors for PU development could then be referred for additional risk assessment. The purpose of a trigger question is to appropriately elicit important information. In this case, the trigger is meant to act as a clue. 14 When the trigger is present, the nurse asking the question is prompted to consider what further actions or questions should be asked and answered to avoid PU development. Our clinical experience strongly suggests that professional nurses are familiar with the use of trigger questions. For example, a nurse may ask an emotionally distressed patient, Do you have thoughts of hurting yourself? A positive response of yes by the patient acts as a trigger that prompts further questions and actions designed to protect the patient from self-harm. The term trigger tool was first used by Classen and associates 14 to identify adverse drug events in large groups of hospitalized patients. Trigger questions have also been advocated for use in identifying sentinel events in the macroenvironment for internal medicine patients. 15 These questions have been clustered into tools for quantifying organizational harm in several different populations 16-18 and for measuring the impact of safety programs. 16, 17 For example, the Trigger Tool for Measuring Adverse Events in the neonatal intensive care unit is used to identify factors associated with an adverse event, such as a death, in the NICU. 18 The ICSI suggests developing trigger questions to identify patients who require further assessment for PU risk. 2 They proposed 3 trigger questions for evaluating PU risk in an outpatient patient population ( Table 2 ). 8 We reviewed these questions and concluded that they also provided a basis for PU risk assessment in our NICU. Nevertheless, subsequent literature review revealed no published studies or psychometric testing for ICSI trigger questions. Adapting ICSI Questions Use in the NICU We noted that the ICSI trigger questions correspond to constructs used within the Braden Scale For Pressure Sore Prevention and Braden Q instrument. For example, the first 2 ICSI trigger questions about moving extremities TABLE 2. Trigger Questions for Pressure Ulcer Risk Proposed by the Institute for Clinical Systems Improvement Is the patient Moving extremities and/or body inappropriately for developmental age? Responding to discomfort in a developmentally inappropriate manner? Demonstrating inadequate tissue perfusion with evidence of skin breakdown? and/or body and responding to discomfort appear broadly similar to the Mobility and Sensory Perception subscales in the Braden Scale and Braden Q. Posed as trigger questions, identification of these factors is intended to prompt additional consideration of the potential adverse effects of intensity and duration of pressure. Similarly, the third ICSI item queries tissue perfusion. Posed as a trigger question, it appears broadly similar to the tissue perfusion and oxygenation section of the Braden Q instrument. Identification of this associated factor is intended to prompt further consideration of issues related to the tolerance of the skin and supporting structures construct. These questions can also be used to trigger additional consideration of factors unique to the NICU. For example, Question 1 Is the infant moving extremities and/or body appropriately for developmental age? is designed to prompt additional consideration of preventive interventions in any neonate whose movements are restricted or prevented by pharmacologic agents for treatment purposes, and any infant whose movement is limited by ventilation, cooling therapies, extracorporeal membrane oxygenation, or a surgical procedure. The second item queries Is the infant responding to discomfort in a developmentally appropriate manner? This question prompts additional consideration for any infant who may have sensory issues such as myelomeningocele, spinal cord injuries, encephalopathies, or other neurologic disorders. The third item queries whether the infant demonstrates adequate tissue perfusion based on the formula (mean arterial pressure gestational age and/ or capillary refill 3 seconds). Identification of this factor prompts additional consideration in any infant who is experiencing diminished tissue perfusion because of various cooling therapies, dehydration, or shock. We also deemed it necessary to change the direction of the questions to better align with current documentation patterns. Our NICU nursing staff chart by exception and identify events outside an established norm ( Table 3 ). To encourage NICU nurses to implement the trigger questions, we embedded them into electronic flow sheets in the section with skin assessments that are assessed by every nurse, on every patient at every shift ( Figure 1 ). A negative response to any of the 3 trigger questions cues the bedside nurse to send a referral to a WOC nurse. These nurse-to-nurse consults do not require physician orders at our facility. The WOC nurse typically provides a complete WON200423.indd 3

4 Schumacher et al J WOCN November/December 2012 TABLE 3. Trigger Questions Adapted for Use in Our Neonatal Intensive Care Unit Is the infant Moving extremities and/or body appropriately for developmental age? Responding to discomfort in a developmentally appropriate manner? Demonstrating adequate tissue perfusion based on the clinical formula (mean arterial pressure gestational age and/or capillary refill 3 s) FIGURE 1. [AQ6] risk assessment, including the Braden Q, followed by a review of preventive interventions implemented by the bedside nurse and any additional suggestions for management as indicated. Evaluation of the Tool The 3 trigger questions were implemented in the summer of 2009. To evaluate the effectiveness of the questions, a complete Pressure Risk Assessment was performed by a WOC nurse using a standardized tool, the Braden Q, in 15 NICU patients. The risk scores of these assessments were then compared to the recorded results of the 3 trigger questions documented by the bedside nurse to determine if the 3 questions triggered NICU patients at risk ( Table 4 ). We noted that the greatest risk for PU development in our population was related to the use of Continuous Positive Airway Pressure devices. Although not incorporated into the 3 trigger questions summarized earlier, the use of Continuous Positive Airway Pressure is incorporated into TABLE 4. Comparison of Results of Trigger Questions and WOC Assessments Patient Story Results of Trigger Questions by RN WOC Assessment #1 term infant admitted to rule out sepsis #2 was a late preterm (36 weeks gestation) infant who needed airway support in delivery room, weaned from oxygen at 3 hours of age #3 term infant admitted for a pneumothorax #4 was a late preterm infant requiring a cardiac workup for a possible heart condition #5 term infant admitted for septic workup #6 31 weeks gestation infant needed nasal continuous positive airway pressure (CPAP) for 4 d #7 premature infant of a diabetic mother admitted for respiratory issues, nurse sent referral as patient was triggered by device assessment. #8 31 weeks gestation infant needing CPAP, referral as patient triggered by device assessment #9 31 weeks gestation infant, weaned to nasal cannula at 3 hours of age #10 low-birth-weight infant, requiring full ventilatory and At risk pressor support #11 extremely low-birth-weight infant, very immature on jet At risk vent/pressor support #12 late preterm infant, diabetic mother admitted for CPAP and surfactant, weaned to cannula within a few hours Risk is related to device. Risk is related to device Very high risk, gel mattress, developmental supports in place Very high risk, gel mattress, developmental supports in place since CPAP was weaned early #13 Immature infant with contractures. Intubated At risk At risk, specialty mattress initiated #14 24 weeks gestation infant on full pressor support, first clinician assessment. At-risk subsequent assessments #15 infant on CPAP, requiring blood pressure support At risk, triggered by blood pressure and device Very high risk, gel mattress, developmental supports in place At risk, addressed pressure relief and support surface WON200423.indd 4

J WOCN Volume xx/number x Schumacher et al 5 routine nursing assessment of the skin and devices ( Figure 1 ). Should other facilities choose to use the questions as is, we recommend that including a trigger for patient devices would be essential. Of the at-risk patients, 1 patient was not correctly triggered by the first assessment provided by the bedside nurse. A root cause analysis of this patient revealed that PU risk should have been triggered by presence of 2 other factors: tissue perfusion and limited mobility. We used this case as an opportunity for additional education about the role of trigger questions and the value of a WOC nurse consultation when risk factors for PU development are present. Following implementation of the trigger questions in 2009, we observed no net increase in the number of WOC referrals per 1000 patients. Nevertheless, our PU prevalence in the NICU remains remained low at 0.01 per 1000 patient days. Comparison of results from the 3 trigger questions and Braden Q scoring by a WOC nurse demonstrated that most infants are correctly identified by the tool, with the exception of those very immature infants who may be at risk for medical device related ulcers. We also observed that NICU nurses began implementing PUprevention strategies before the initial consultation with the WOC nurse. For example, we have observed that NICU nurses tended to place infants who were unable to move owing to the use of high-frequency oscillator ventilation or pharmacologic induced paralysis on viscoelastic pressure-reducing mattresses. Summary Questions designed as clinical triggers may be used in daily clinical practice to identify patients at risk for various facility-acquired conditions including altered skin integrity. We implemented a 3-item trigger tool to aid NICU nurses identifying neonates at risk for PU development. While these questions do not quantify risk, we have found that they are an efficient initial screening tool when combined with additional assessment and management in consultation with a WOC nurse. KEY POINTS Clinical triggers can serve as flags to aid nurses in identifying patients at risk for facility-acquired conditions including PU development. Based on our clinical experiences, NICU patients are at risk for PU development and especially ulcers related to the use of medical devices. A 3-item trigger tool aided NICU nurses to identify patients at risk for PU development and prompted consultation with the WOC nurse in these selected cases. References 1 Centers for Medicare & Medicaid Services. Eliminating serious, preventable, and costly medical errors never events. http:// www.cms.hhs.gov/apps/media/press/release.asp?counter 1863. Published 2006. 2 Institute for Clinical Systems Improvement. Health care protocol: skin safety protocol: risk assessment and prevention of pressure ulcers. http://www.icsi.org/ skin /pressure_ulcer skin_safety_protocol risk_assessment_and_prevention_of protocol_.html. Published 2007. [AQ3] 3 Institute for Healthcare Improvement. 5 Million Lives Campaign. Getting Started Kit: Prevent Pressure Ulcers How-to Guide. Cambridge, MA : Institute for Healthcare Improvement ; 2008. http://www.ihi.org. 4. Titler MG, Kleiber C, Steelman V, et al. The Iowa Model of Evidence-Based Practice to promote quality care. Critical Care Nurs Clin N Am. 2001 ; 13 ( 4 ): 497-509. 5. Bergstrom N, Alma R, Alvarez D, Bennett MA, Carlson CE, Frantz R. Treatment of Pressure Ulcers. Clinical practice guideline number 15. AHCPR Publication No. 95-0652. Rockville, MD : US Department of Public Health and Human Services, Agency for Healthcare Policy and Research ; 1994. 6. Butler C. Pediatric skin care: guidelines for assessment, prevention, and treatment. Dermatol Nurs. http://findarticles.com/p/articles/ mi_hb6366/is_5_19/ai_n29388272/. Published January 19, 2011. 7. Bergstrom N, Braden B, Laguzza A, Holman A. The Braden Scale for predicting pressure sore risk. Nurs Res. 1987 ; 36 ( 4 ): 205-210. 8. Baharestani MM, Ratliff C ; and the National Pressure Ulcer Advisory Panel. Pressure ulcers in neonates & children: an NPUAP white paper. Adv Skin Wound Care. 2007 ; 20 ( 4 ): 208-220. 9. Noonan C, Quigley SM, Curley MAQ. Skin integrity in hospitalized infants and children a prevalence survey. J Pediatr Nurs. 2006 ; 21 ( 6 ): 445-452. 10. Huffines B, Lodgson MC. The Neonatal Skin Risk Assessment Scale for predicting skin breakdown in neonates. Issues Compr Pediatr Nurs. 1997 ; 20 : 103-114. 11. Curley MA, Razmus IS, Roberts KE, Wypij D. Predicting pressure ulcer risk in pediatric patients. Nurs Res. 2003 ; 52 ( 1 ): 22-31. 12. Anthony D, Papanikolaou P, Parboteeah S, Saleh M. Do risk assessment scales for pressure ulcers work?. J Tissue Viability. 2010 ; 19 ( 4 ): 132-136. [AQ4] 13. Kottner J, Hauss A, Schluer A, Dassen T. Validation and clinical impact of pediatric pressure ulcer risk assessment scales. A Syst Rev Int J Nurs Stud. http://www.journalof nursing studies.com/ article/s0020-7489(11)00218-5. Published July 13, 2011. Accessed May 18, 2012. 14. Classen DC, Pestotnik SL, Evans RS, Burke JP. Computerized surveillance of adverse drug events in hospitalized patients. JAMA. 1991 ; 266 ( 20 ): 2847-2851. Trigger tools have been used to enhance safety. [AQ5] 15. Richter ED, Barach P. Occupation and environment in internal medicine: sentinel events and trigger questions. Mt Sinai J Med. 1995 ; 62 ( 5 ): 390-400 16. Griffin FA, Resar RK. IHI Global Trigger Tool for Measuring Adverse Events. 2nd ed. IHI Innovation Series white paper. Cambridge, MA : Institute for Healthcare Improvement ; 2009. http://www.ihi.org. 17. Takata G, Mason W, Taketomo C, Logsdon T, Sharek PJ. Development, testing, and findings of a pediatric-focused trigger tool to identify medication-related harm in US children's hospitals. Pediatrics. 2008 ; 121 ( 4 ): e927-e935 doi:10.1542/ peds.2007-1779. 18. Sharek PJ, Horbar JG, Mason W, et al. Adverse events in the neonatal intensive care unit: development, testing, and findings of a NICU focused trigger tool to identify harm in North American NICUs. Pediatrics. 2006 ; 118 : 1332-1340. WON200423.indd 5

AUTHOR QUERIES TITLE: Development of a Pressure Ulcer Trigger Tool for the Neonatal Population AUTHORS: Bette Schumacher, Mary Askew and Kathy Otten [AQ1] Please provide professional titles/designations for all authors. [AQ2] Please provide the e-mail address of the corresponding author. [AQ3] For refs. 2, 3, 6, and 16, please provide the access date. [AQ4] Reference 12 has been updated per PubMed. Please verify. [AQ5] Should the sentence (Trigger tools have been used to enhance safety.) be deleted or set as a footnote in the text (where ref. 14 is cited)? [AQ6] Please provide a caption for Figure 1. WON200423.indd 6